aboutsummaryrefslogtreecommitdiff
path: root/freq.pl
diff options
context:
space:
mode:
Diffstat (limited to 'freq.pl')
-rw-r--r--freq.pl147
1 files changed, 147 insertions, 0 deletions
diff --git a/freq.pl b/freq.pl
new file mode 100644
index 0000000..f5feb6a
--- /dev/null
+++ b/freq.pl
@@ -0,0 +1,147 @@
+#!/usr/bin/perl -w
+
+sub parse_matr {
+ my $fn=shift;
+ my $l;my @l;my $k2;
+
+ open FM,"<$fn" or return;
+
+ $l=<FM>;
+ my @k1=split ' ',$l;
+ $i=0;
+
+ while ($l=<FM>) {
+ @l=split ' ',$l;
+ $k2=shift @l;
+ for ($j=0;$j<=$#k1;$j++) {
+ $matr{$k2}->{$k1[$j]}=shift @l;
+ }
+ }
+ close FM;
+}
+
+sub compact {
+ my @r;
+
+ @_=sort @_;
+ push @r,shift @_;
+ while (@_) {
+ if ($_[0] eq $r[-1]) { shift;next }
+ push @r,shift @_;
+ }
+ return @r;
+}
+
+sub save_matr {
+ my $fn=shift;
+ my $l;
+ my @k1;my @k2;
+
+ open FM,">$fn";
+
+ @k2=sort keys %matr;
+ @k1=();for (@k2) {@k1=(@k1,keys %{$matr{$_}})};
+ @k1=compact(@k1);
+
+ print FM " @k1\n";
+
+ for ($i=0;$i<=$#k2;$i++) {
+ print FM "$k2[$i]";
+ for ($j=0;$j<=$#k1;$j++) {
+ print FM " ",defined($matr{$k2[$i]}->{$k1[$j]})?$matr{$k2[$i]}->{$k1[$j]}:0;
+ }
+ print FM "\n";
+ }
+ close FM;
+}
+
+%folds=(
+ '~' => '`',
+ '!' => '1',
+ '@' => '2',
+ '#' => '3',
+ '$' => '4',
+ '%' => '5',
+ '^' => '6',
+ '&' => '7',
+ '*' => '8',
+ '(' => '9',
+ ')' => '0',
+ '{' => '[',
+ '}' => ']',
+ '"' => '\'',
+ '<' => ',',
+ '>' => '.',
+ 'P' => 'p',
+ 'Y' => 'y',
+ 'F' => 'f',
+ 'G' => 'g',
+ 'C' => 'c',
+ 'R' => 'r',
+ 'L' => 'l',
+ '?' => '/',
+ '+' => '=',
+ '|' => '\\',
+ 'A' => 'a',
+ 'O' => 'o',
+ 'E' => 'e',
+ 'U' => 'u',
+ 'I' => 'i',
+ 'D' => 'd',
+ 'H' => 'h',
+ 'T' => 't',
+ 'N' => 'n',
+ 'S' => 's',
+ '_' => '-',
+ ':' => ';',
+ 'Q' => 'q',
+ 'J' => 'j',
+ 'K' => 'k',
+ 'X' => 'x',
+ 'B' => 'b',
+ 'M' => 'm',
+ 'W' => 'w',
+ 'V' => 'v',
+ 'Z' => 'z',
+);
+
+for (values %folds) {
+ push @folds,$_;push @folds,$_;
+}
+%folds=(%folds,@folds);
+
+sub fold_char {
+ if (exists $folds{$_[0]}) {
+ return $folds{$_[0]};
+ } else {
+ return undef;
+ }
+}
+
+sub upd_matr {
+ my $fn=shift;
+ my $l;
+
+ open FM,"<$fn";
+
+ while ($l=<FM>) {
+ for ($i=0;$i<length($l)-1;$i++) {
+ $j=fold_char(substr($l,$i,1));
+ $k=fold_char(substr($l,$i+1,1));
+ if (defined($j) && defined($k)) {
+ $matr{$j}->{$k}++;
+ }
+ }
+ }
+ close FM;
+}
+
+$MATRFN='/tmp/freq.matr';
+
+parse_matr($MATRFN);
+
+while ($f=shift) {
+ upd_matr($f);
+}
+
+save_matr($MATRFN);